Hidden Value, Democratizing Data Analytics, and Night Shift Development with Tim Tutt

Callye Keen and Tim Tutt navigate the intricate landscape of data analytics. Highlighting the latent promise of expansive datasets, they illuminate how businesses, especially within tech and government, can capitalize, monetize, and unearth valuable insights from the data that often goes unnoticed.

Topic Highlights:

00:00 - Introducing Tim Tutt

Callye Keen introduces Tim Tutt, CEO and Cofounder of Night Shift Development, helping organizations operate more efficiently by building them the tools they need to better exploit their data.

[06:11] - The Data Deluge: Navigating the Surge of Modern Data Acquisition

Callye Keen highlights the evolution in data acquisition, emphasizing the ease with which modern sensor systems can capture data. He underscores the importance of refining the data storytelling process amidst this data influx, stressing the challenges of sifting through the vast amounts of collected data to extract meaningful and actionable insights.

[10:47] - Edge Computing in Action: Chick-fil-A's Real-time Data Mastery

Tim highlights the growing significance of edge computing by giving a fascinating example of Chick-fil-A, which runs one of the world's largest Kubernetes clusters. Each of their locations independently analyzes data for optimal cook times, customer influx, and revenue predictions. This real-time, localized data processing exemplifies how businesses, and even the defense sector, can benefit from analyzing data right at its source rather than solely relying on central aggregation.

[26:59] - Tailored Digital Experiences: Instagram's Model

A look at how platforms like Instagram are reshaping user experiences through personalized content, thanks to behavioral data analytics.

[28:20] - Telemetry Sensors and Automotive Advancements

Tim emphasizes the role of data in automotive innovations, focusing on how Tesla's telemetry sensors are revolutionizing driving safety and car efficiency.

[32:00] - The Government's Data-Driven Decision Making

The conversation shifts back to government agencies, where data analytics is being used to refine contract language, assess proposal viability, and streamline operations, ensuring taxpayer value.

"Data is the new oil and a lot of people are collecting data that they just aren't doing anything with... if you got data, you want to be analyzing it because there's going to be some nugget in there that you can use." - Tim Tutt

Callye Keen - Kform

https://kform.com/ 

https://www.linkedin.com/in/callyekeen/ 

https://youtube.com/@kforminc  

https://twitter.com/CallyeKeen 

Tim Tutt - Night Shift Development, Inc

LinkedIn - Tim

LinkedIn - Night Shift Development 

ClearQuery.io

Nightshift Development

https://twitter.com/nsdtechio 

About Tim Tutt

Tim is currently the CEO and Cofounder of Night Shift Development, Inc., which focuses on helping organizations make sense of their data. Using his background in natural language processing, machine learning, and data analytics, he created and productized a solution aimed at allowing nontechnical users to have conversational experience with their data.

Transcript

Speaker 1: 0:00

Instagram has my number in terms of targeted ads. I cannot tell you the number of things that I have bought off of Instagram ads. It's because they've collected enough data in terms of my behaviors using the app, the things that I interact with, the things that I am passing over, and they're able to use that to say OK, great. Not only can we send the right type of content to Tim, we can also send the right types of ads where Tim's most likely going to buy. Even though we've got a number of commercial organizations that dealing with massive amounts of data, government's still pretty far ahead of the amount of data that they're collecting.

Speaker 2: 0:38

Welcome to the Startup Defense. My name is Callie Keene. Today I have Tim Tut. Tim is an expert at data analytics, has a lot of experience in the government space, showing organizations how best to use their data, unlocking value. And, honestly, tim, I'm not the expert here. I am here to learn and I'm really excited for this conversation for that reason, before we dive into the technical details, I'd like the audience to get a good feel for you. What's the story of Tim? What are you passionate about right now?

Speaker 1: 1:11

Sure Well, hey, thanks so much, callie, for having me on Excited to be here and chat with you. So my background I come from the government space. I spent the last 15 years of my career operating inside the US intelligence community, really focused on how we do large scale data exploitation and search and discovery. You know, that's kind of where I started. You know, from a career background, I played Little man Between my End Users and their data.

Speaker 1: 1:38

I was the guy that people would come to and ask questions and say, hey, we're looking for a particular individual, particular location, particular patterns or things of interest that we can find to take the next action, take the next step, put X on the ground wherever we particularly need to go, and I'd go and write queries, run those against a massive supercomputer and come back with answers for our end users. And that became a bit of a repetitive cycle. That's kind of one of the things that prompted me to launch out and start the company that I run now, night Shift Development, where we're really trying to help democratize access to data and data analytics. How do we help enable people to get value from their data without needing to have this deep technical background A large part of the way we do. That is with our core product, clear Query, which is designed to make it super simple. We kind of joke that it's, you know, siri meets your data, but you know it's making analytics so simple that anyone can do it, no matter what their technical skill level is.

Speaker 2: 2:43

That's fantastic. I've got a number of friends in commercial data just call it that and played around with R and some of the tools that are available in that space, and I have a little bit of a coding background, definitely technical background, so pretty high, I'd say, barrier to entry business to get into, and so I think a lot of companies, whether they're commercial or defense, they also don't understand the value of the data that they have or the best way to solve a problem, because it's often counterintuitive what needs to be measured or where the data could be inferred. Not only is the tools fairly complex, so it requires, you know, some skills and some training, and then the underlying math, of course, can be probably difficult as well, right, but then there's a creative element to it thinking through that problem. Well, what kind of data do I need? Have you seen that creativity really play out in the space?

Speaker 1: 3:39

Yeah, you know, this is one of the interesting things, I think. In the government space in particular, I found that we're collecting mass amounts of data. There's usually lots of interesting data laying around somewhere being collected, siphoned off into some silo, and usually it's this big challenge of figuring out what that data is and how it melds with other data. I think you're right that in a lot of organizations you do have groups of people that don't even understand what data they need to get started. They know they need data. Everyone's kind of heard this data's a new oil we need to be drilling in, we need to be exploiting it in a better way. But not everyone really knows how to capture that data, clean it, which is another big factor. When you're talking about doing data analysis, data exploitation in general, you're making sure that data is engineered in a way that you can actually even get value from it. But I think the other thing that you do tend to find is people that understand their business, that understand what it is that they're driving at in general, whether that's commercial or defense space, those subject matter experts. They know the right questions to ask. They may not know how to get the data. They may not know where that data comes from. But they do know that. I know these are the questions I need answers to and I know they're going to be follow-ups. I know this is going to be a conversational, iterative process for me to get the right value out of it.

Speaker 1: 4:59

A lot of times, what we end up doing when we first come into an organization it's A helping them understand what data they have. B helping them get that cleaned and engineered in the right way. And then you know, clearquerie is there for that analysis piece. The long tail of it is getting to that data storytelling aspect. How do we then turn that into actionable insights so that we can make the right decision and make the right investments that we need to, and figure out what the next best action is and present that to the right decision makers at the end of the day, and that's a lot of what we try to do as an organization. So our core product, clearquerie, helps to solve a lot of those problems within the government space, within commercial space. You do have these earlier parts of this data story, like you just mentioned finding the right data and then be you know, making sure it's set up for you to even do that analysis in the first place.

Speaker 2: 5:52

Yeah, this is very key. This is a good time to have this conversation because we're building next generation sensor systems. We're taking in a lot of data and in some cases, more data than you'd think per second would even live on a computer and then we're trying to ingest and augment and move that data, transport it in a number of places. It's incredibly important what you're talking about and I'd like to unpack that a little bit as much as makes, let's say, public sense.

Speaker 1: 6:19

Sure.

Speaker 2: 6:19

I'll just give a very easy commercial example. It's like I've been developing products for 15 years. We've been making things my entire life and 10 years ago to make a new sensor system, we might go through spins and spins and in knowing exactly what we were going to capture, was so critical knowing how we would use that data. Later we might have some discover ability. But now it's very easy to just deliver a kit where I can pull in any type of data, any type of sensor system, and reconfigure it and pull in so much. You know, like Putting out an arduino out into environment and just start siphoning in environmental data or whatever have you. This is we're just getting started in data acquisition and then in ingesting it at edge. But then this is gonna put so much more pressure on you and night shift, because as it's easier to acquire data or build the system, so acquire data. That story has to be more and more refined because you're just making more of a mess versus getting intelligible information.

Speaker 1: 7:28

You're not wrong, and you know this is one of the things that we tend to look at broadly is it's a state of triage process, right? So data analysis, data exploitation is also. It's also about this triage. How do I filter down to the data that actually matters? You know things that I tend to kind of really look at, especially when you move into the world of streaming data or data edges. You know what types of capabilities can we also put at the edge to filter that data in a meaningful way, so that we only look at the things that matter?

Speaker 1: 7:57

Sure, you're collecting lots of data. Maybe it's data coming from wifi networks or data coming about, you know, weather sensors. What actually matters in that data for the type of analysis we're trying to do? Do we only care about major events? You can actually think about this from a cyber security standpoint.

Speaker 1: 8:12

Alert fatigue is a major issue with most of the cyber security platforms that you have out there. I've got thousands and thousands of the words coming out, but most of them don't actually matter. Most of them are just, you know, standard. Hey, you know this is activity happening, but it looks anomalous because the systems haven't really refined themselves, that haven't really fine tuned to a point where they can really only highlight the really interesting bits for you. Same thing, whether it's you know a sensor that you're collecting data on. Maybe I only have certain targets of interest that I care about. Let's whittle down to those. Or maybe I hear about those targets of interest in their secondary networks.

Speaker 1: 8:50

So, from the overall data analysis perspective, there's a handful of things that you want to look at. It's you know the streaming data, how we make sure that we're only capturing and analyzing the data that we really really care about, and everything else maybe we want to store for later, just in case you want to. You know, do some other deep analysis on that. For broader atmospherics, not everyone needs everything. You only. You only need you know the small universe of the world that you care about it any Any given time, and I think this is a thing that we have to get better about as we're building these solutions not just collecting the data, but helping people to find the solution. So triage that data up front before they even start doing their own analysis on it.

Speaker 2: 9:34

Yeah. So this is a good. It's a good avenue to go down on a conversation of data transport and whenever you transport something, we're talking about bandwidth, but we're also talking about security issues. So the classic example of this is I can send hd Video all the way, backhaul and transport wherever it's gonna go.

Speaker 2: 9:55

Or I can just say, hey, I need a bounding box around x. You know, I already have gis information. I don't need a hd video stream of this thing. Or do I write, or do I need to store it and do further actionable information. But until they work with somebody like yourself and understand what the data is right now, it's how many hd streams of video can I ingest into a Tactical edge computer and how fast can I transport that data, and I see that as a security issue. But that's from my background. I see this security issue but it's also an intelligible or actionable data issue. So if they talk to you and say I need this person right at the edge to have this data as fast as possible, we isolate just what's the smallest increment of that to take action on and then worry about all that other funds.

Speaker 1: 10:47

Wait? Yeah, that's exactly right. Edge computing is becoming more and more of a hot topic. Right, you're starting to see that more the government space.

Speaker 1: 10:54

Every now and then you'll see rfi's come out, but I didn't even say in the commercial spaces. You're seeing that chick filet, for instance, they launch I think there's a blog post a while back says they run the largest kubernetes cluster in the world and it's because they have these individual pods deployed at every chick filet location. But each one of those locations is doing its own analysis and can tell you down to the second hey, here's perfect amount of cook time that you need to make sure this chick is your nice and crispy, etc. And the number of individuals that are coming in that store, what their revenue is, and they can do all the prediction at the edge. Meanwhile, they aggregate that date on the back end for all of the stores to determine globally how they want. That's fascinating and that's actually what we should be doing.

Speaker 1: 11:41

You know, on a defense side as well, we are operating so many different a or is and we care about the small portions of the world and how. What's happening there with the data that we're collecting in that localized environment. How do we then help enable those individuals at that edge to get the analysis that they need very, very quickly and then ship back all the data later for the broader global analysis. Right? How do I make sure that at a border crossing we can identify someone that's concerning and of interest in a particular arena but also gather all that data later for broader analysis when we're looking for interesting ways to get into networks across the board there?

Speaker 2: 12:23

We could probably extend this chick full a conversation and come up with some interesting parallels. But you're turning that to machine vision and say their drive through and it's like okay, well, at edge, we wanna do plate scrubbing because we don't want that in that data. We wanna do head count, we wanna do instance count so we can make predictive analysis of, okay, this, how many people we need out there with their little clipboards like they do? But also, I need immediate, actionable information to headquarters of like ever in conditions. Right is like okay, this took a really long time, the dwell time was really high, or somebody moved in a way that's not normal. What was the issue? Or somebody had a car accident there. What's the issue?

Speaker 2: 13:07

This is why we launch this podcast, because you'll see people do interesting things in commercial markets, because there's immediate dollars to be made there. If you wanted to implement that solution, well, chick-flea's already done it. But if you wanted to implement that solution, there's AI boards right now where they could do right out of the box, open source, plate scrubbing, cv of cars, car counting, face grabbing, sentiment analysis on the faces all kinds of crazy stuff. It's actually not that difficult to do anymore. What we see is every time that we have this condition or somebody that wants to do something. In commercial there's a 10x need. In defense we have all the same problems, but in commercial, if they wanted to roll out and say we want more computing power, we're just going to put it out on Chick-Flea's they're all domestic. It's not going to be that big of a deal to just upgrade systems or upgrade their servers, whereas defense that might be absolutely impossible.

Speaker 1: 14:05

Yeah, and it becomes more challenging especially when you talk about mission critical areas where we have low access to begin with. Hey, maybe even getting the thing into a particular location was hard to begin with, let alone getting something updated, even if that's transporting across the network, because you don't want to become alerting in an environment where it's particularly dangerous, where you could be caught.

Speaker 2: 14:30

Chick-Flea is not worried about highly contested environments. We have a Wi-Fi and GPS denied fast food. It's not really an issue. But that's an everyday upfront issue that we would deal with. And then it really just comes down to the best thing is when there's no data at rest and we're transporting as little as possible. So it's like knowing what data we should use as upfront is critical.

Speaker 1: 14:53

Absolutely, and this is where AI and machine learning capabilities at the edge become more and more critical. And then, on top of that, making those AI and machine learning capabilities small enough, compact enough that you can deploy at the edge, because sometimes, if you're talking about an Arduino, if you're talking about small interface, you don't have much storage space to begin with, much compute power to begin with. Can I actually run this algorithm on the device that I am deploying at the edge? That becomes a major consideration. So I think you're exactly right. You know, a lot of the commercial problems do get 10xed in the government space, especially when you start factoring in the security concerns and security controls around it, but then also the other constraints that we have to operate in. Then you factor in the who's, the user of the device.

Speaker 1: 15:44

I'm not always going to have data scientists, say, to engineer at the edge. I need to be able to help enable my operator to be able to use this thing and then also handle the failures on the back. That's one of the biggest challenges that you know we run into. Hey, this thing breaks. I don't have time to wait for someone to kind of walk me through it. I need to either be able to fix or move on, because we have, you know, things to do and we have to need to move on or it puts us in a bad position. So, understanding those challenges this becomes more important the more dangerous emissions that is. But those things do occur and exist and that's where it's most critical for us to have these technologies available at the end of the day.

Speaker 2: 16:24

Good design needs good constraints.

Speaker 2: 16:27

So, we're headed towards good design, I suppose. Hey, I wanted to ask you because I'm on your LinkedIn and I saw that you have worked with decode and so we've had Meg Vorlin on the show I've mentioned to you. I'm a big fan of what decodes been able to do and it fits kind of my dual use thesis is like, okay, if we start as a defense company, we should look at how we can scale with commercial applications. If we're thinking about a commercial company, we should think about how we can serve our country. Think of, like what's my first customer?

Speaker 2: 17:00

Because, you know, in startup world they're like Find that beach head customer, like what better beach head customer? And something is actually good. You know, you know, actually breach of beach head. You know, yeah, you know, like come on, like let's be serious about this. But no, they like okay, early adopters would be willing to cut a check for your thing to exist. Early adopters would be willing to co-develop something for you, or they have a 10x problem to be had. And I think in every case that I've come across there really describing dual use, yeah, I just wanted to dive into that. You know your interaction with decode and that experience how's that been for you?

Speaker 1: 17:38

When I work for decode for a long time in a couple of different capacities, so happy to dive into that. Big shout out to Meg Borland, who you mentioned, and Megan Metzger and Rebecca Devault, who are, you know, over there running different parts of you know that organization. So you know decode score mission right is to help bring commercial technologies to government. But also they kind of have a dual mission, which is how do we help government understand how to identify and bring in the right technology, which I think is a very critical and missing piece. You know oftentimes government for the longest time the government was the most innovative place for new technology. We are doing things that you don't see in commercial spaces, and that's still true in a lot of respects. But as data becomes more prevalent in the commercial sector, as people become a lot more Interested, there's a lot of innovation happening in the commercial world that the government needs to be able to leverage but oftentimes moves way too slow to leverage and this becomes a big issue, especially for smaller companies that don't understand government buying cycles, that have investors that are pressuring them to drive and get revenue quickly and don't necessarily know how to move through the wickets of government. So what decodes doing is basically connecting the government individuals that, hey, we know we need technology, we need innovative things. We still know how to maneuver through the process and wickets to get this done faster or fast enough, and then, on the flip side, commercial companies that are driving in that have innovative tech and want to be business with the government.

Speaker 1: 19:15

We started working with decode Because I was back in 2020 when we launched our sales team and most of my sales team had only sold commercially, so wanted to help them understand how to run through the buying cycle, so we got your team kind of spun up there.

Speaker 1: 19:31

Since then, I've also become a limited partner with decode capital, so I'm on the other side helping to invest in companies that are trying to get into the government space. For me, that's hyper exciting because, having worked in the government for a long time, I can see a lot of interesting use cases that you know. I want to help plug these companies into that. I wish you know we had in the past Yep. So, yeah, I jumped in as an LP with decode capital and I've been really excited to work with some of the companies coming in helping to drive that because, looking back in my days on the government side, bringing in some of that tech things that I wish we had before, things that I wish we could actually do something with. It's nice to see that actually working out in a good way for us.

Speaker 2: 20:13

We had a number of VCs on the show. You know all different sizes angels. We had some of the team from shield capital on. This is very interesting because we had people from AAL and different organizations that they've gone from working in the IC or do they've gone through acquisition it kind of seen that whole rigmarole and then they've gone in and maybe been a contractor and then they've become a part of the whole startup world or VC piece. And so I love what you're saying, because it's a common through line is to me. I spend my whole life in defense making things for defense. It's still kind of a mystery to me. You know you learn something new or things evolve, like they've evolved in the past few years about with DIU and afwarks and the different programs, and really evolved even how SBIR's work is evolving right now, and so there's a lot to learn, is all I'm saying. It's not as easy as like I get my LLC and put out my digital signal and start working. So your experience is invaluable for these startups.

Speaker 1: 21:20

Yeah, absolutely, and I think that's one of the biggest things that I really love about it is there's so much cool tech out there, but unless you understand the government use cases, it becomes really hard for you to sell that to the government. This is one of the things I love about Dico. It is they do really help organizations to hear that out along the way. Here's how we help you find the government use case under sell to government, because selling to government isn't about, hey, how many dollars are we going to make at the end of the day. You do have other factors that matter. Hey, how do we save time on this particular mission? How do we help find an answer that isn't easily findable, something that's a needle in the haystack?

Speaker 1: 22:02

Even though we've got a number of commercial organizations and dealing with massive amounts of data, government still pretty far ahead of the amount of data that they're collecting, and the thing, the needles that they're trying to find are much more interesting and much harder to find. So how do we drill in and deal with that? Then, on top of that, you know how do we handle the obscure security issues that the government has to deal with that most organizations don't ATOs, fed ramp, etc. You know that's a whole separate process. If you have no idea what any of those letters mean, that's a. It's a different piece. So you know Dico is doing a very good job of helping organizations to drive into this space in a very effective way and helping innovative technology companies bring those capabilities to the government space to solve some of the most interesting and important problems that we face as an Asian.

Speaker 2: 22:55

I was joking, because tech likes to develop its own jargon, use words in a different way. But I'm also in manufacturing, so manufacturing has its whole own universe of words. But then defense is the king of acronyms and jargons. And somebody says, oh, this is like yeah, ato authority to operate CSO, then Palm, and then this. And you're like is palm an agency or is it a? You know, it's like, okay, like what is this, what is that?

Speaker 2: 23:21

And so it's this whole language that people need to get up to speed, and I think the old method of this was Well, I'm gonna have to go and hire somebody from do d and put them on my advisory team or put them in the C suite, because they know all this vocabulary and they kind of know how to navigate. But now we have organizations, community organizations, venture organizations and then new now and I think really developing these innovation organizations within the community, like within DoD, even within the IC, reaching out and saying like, hey, we're going to help you get in here and figure out what to do, where to go, who to talk to, what's important.

Speaker 1: 24:01

Absolutely. And yeah, and you're starting to see that a lot more. It's a thing that the government's been lagging behind on. Yeah, you're right, even certain IC organizations. Cia now has a group that is designed to go out and find innovative technology companies. They are the front door for anyone wanting to bring interesting technology to the agency. It's typically getting the agency. Unless you know that space and unless you really know, you're not getting it. There's a high barrier for entry, for good reason. It absolutely is something that, hey, there are still things that we need and want to take advantage of. How do we do this in an effective way?

Speaker 2: 24:36

Yeah, you don't want everybody pitching in Qtel All the time.

Speaker 2: 24:40

I have the slightly important things to do, other than every inventor saying I've got a cool idea. Listen to me, I want to be a good advocate for innovators that are listening to the show and help them understand what you do and why they're making a tech product. Let's say that I have a new sensor product or I have a new compute product and I'm listening to my customer's mission profiles or their user story. This is what we're thinking about doing. Why should they double down in thinking about how to use data?

Speaker 2: 25:13

There's a company that commercially came out. They had an app. Their app failed and this is in the DC region. Their app failed but they realized that they had created a great way to pull data and their company shifted over to doing just that, because the purpose of their app was no good, but what they were able to do with the data that they were pulling from their app was so valuable that it could be acquired. I think that when you say data is the new oil, a lot of people are just like Jeb Clampett or something. They're just walking around. They don't know that they're sitting on this giant wealth of potential value. I want to just talk about that really quickly, because this is something I very, very rarely see people thinking about how can I add additional value with my products or my systems through?

Speaker 1: 26:01

data. Sure, now, that's a fantastic point and you're right, there are a number of groups out there that collect a lot of data that they don't even do anything with and they really could, because that could be its own business line, if not its own company. It's an interesting company I was talking to maybe a month and a half ago where they built a new chatbot interface, and this chatbot interface is designed using chat, gbt and a couple of things, but it's designed to mimic a friend, and what was interesting about it was they launched this platform and ran a couple of ads, got a bunch of folks using it it's kind of a testbed and had hundreds of thousands of messages over the first couple of days within the app of. Hey, here are things that people are asking and here's how the chatbot's responding. Here's what they're asking, and I'm like that's your value right there. Your value is the data that you're collecting on what people are asking, what they're interested in. You can take that and you can start to derive really interesting things from it. Hey, here's some interesting buyer personas, people that are interested in these things. Here's something that we can target to an individual. This is one of the things I talk about.

Speaker 1: 27:11

Often, when I talk about data, I look at Instagram. Instagram has my number in terms of targeted ads. I cannot tell you the number of things that I have bought off of Instagram. It's not Amazon for me, it's Instagram. I'm guilty as well. I'm guilty, but it's because they've collected enough data in terms of my behaviors using the app, the things that I interact with, the things that I am passing over, and they're able to use that to say okay, great, not only can we send the right type of content to Tim, we can also send the right types of ads where Tim's most likely going to buy, or he's at least going to click on these ads, which is where Instagram generates their revenue at the end of the day. And now, all of a sudden, they build an even better profile for me of hey, these are things that Tim really cares about. That kind of targeting and user behavior is hyper interesting.

Speaker 1: 28:05

But then you think about it from all these other perspectives. Let's say you're talking about sensors on a car or a vehicle. What are the driving patterns of an individual? How can we increase the safety and the mechanics of this vehicle so that it drives a lot more effectively? I think Tesla and all these cars that now have all of these telemetry sensors built in them that are being shipped back to the mothership. That is going to help us develop and build more effective cars or more safe cars for the consumers on the road. So there's a million different ways that you can leverage even just users behaviors of the application or the sensor that you have to derive new data that becomes really interesting to leverage in other perspectives.

Speaker 1: 28:51

Then you think about that from government perspective as well, even looking at ad data where people are in their time, where they are, how do we get to them in the most effective ways? It becomes a really, really interesting sea for people, no matter where you are. So, yes, data is the new oil and a lot of people are collecting data that they just aren't doing anything with. That's what one of the things that we want to do is help enable them to get value from that data or at least start seeing interesting trends without needing to know much about how to analyze it. Throw your data into our platform and it will help you identify some of those insights. But, yes, absolutely, it's just a key goal of if you got data, you want to be analyzing it, because there's going to be some nugget in there that you can use for, whether that's more profits or a better improvement of an overall system. Whatever that is, there's a lot of different ways to leverage it in an effective way.

Speaker 2: 29:45

I'll give two examples. One, when I've helped commercial startups this is common in e-commerce. It's like, hey, I'm going to use some different trend analysis tools. I'm going to go look at Google searches and I'm going to go look at who actually cares about this. So, yeah, you could look at hashtags, you could look at the keywords that are built, or just content on YouTube. So like, are people actually talking about the thing that I'm going after? And hopefully that graph looks like it's going up and to the right, that people are caring about this more and more.

Speaker 2: 30:15

We actually use this to find good government problems as well. But just on the, what kind of SBIRs are getting funded or what kind of programs are people putting out there? Or what kind of pitches were accepted in XYZ events? Or what topics were people speaking about at XYZ industry event or trade organization? Like, who's speaking and what are they talking about?

Speaker 2: 30:37

So we can infer problems and stories from this, and what I want to pull from that is my second example is I look at that and I think, okay, well, if I'm building, I have a little wireless retransmission device, right, we put accelerometer and GPS on it. It's like, well, why do that? Well, because there's a ton of things that you can get in for data. Also an accelerometer. I can know quality of transmission. If you don't want it, you don't want it.

Speaker 2: 31:08

I know from looking at those stories who's talking about what is. There's a lot of potential applications for that data. I need to go out and talk to somebody like yourself, or I need to go and look at the speeches at those conventions and say, okay, well, I have this ability to gather additional data. What can I do with this thing that I have that I'm not thinking about. I just want to double down on this because I've seen this repeatedly. It'd be a problem is somebody's creating a product and it's okay, but from a data perspective it's excellent. They're just not monetizing or selling that as the actual product itself.

Speaker 1: 31:45

Absolutely. I'm just kind of double down on your point. Yes, there's a lot of different interesting. I love somebody kind of the use cases you were talking about looking at. One of the things that we look at, for instance, is USA spending data every now and then, because then you can start to see trends of organizations that are investing in interesting technologies.

Speaker 1: 32:02

But on the flip side, there are a lot of government organizations that are looking at that and saying are we spending too much with a particular vendor, with a particular subcontractor? Are we spending too much in a particular arena? Because if we are, hey, if we have eight different organizations buying the same type of technology, should that be consolidated somewhere or is it actually needed? Or is there a better way for us to buy and procure these things? Are there other interesting things that we can look at there? Is there a particular language and contracts that we should be extracting and evaluating for how we go after business with the government? But also from the government's perspective, are there things that they need to be looking at and proposals that seem to be high indicators of lots of fluff or a really successful contract to help them determine how to actually move forward? Yeah, there's a lot of different ways that you can leverage data in a very effective way here.

Speaker 2: 32:58

So a lot to think about. Tim, I really appreciate you taking on time to be on the show. It's really been great. I kind of have a mental checklist of homework for myself right now. I appreciate your time.

Speaker 1: 33:08

Absolutely Well. Hey, I really appreciate you having me on. It's been a fun chat for sure.

Speaker 2: 33:13

This has been the startup defense.